mindspore2022

584 MB

Tree: f27a16c8f8

Author	SHA1	Message	Date
looop5	2fcf970f69	deal with workspace deal with workspace on cuda	4 years ago
zengzitao	8dcff8d83c	refactor tile op and in expander open on gpu	5 years ago
r1chardf1d0	3b32995936	enable stitch fusion on bert	5 years ago
dayschan	3c6c30024c	dump graph_kernel_split info	5 years ago
dayschan	08345c54ea	[GraphKernel] Replace Assign with InplaceAssign 1. added a pass to replace Assign with InplaceAssign. 2. bugfix in eliminate_redundant_output. the side-effect node should not be eliminated. 3. bugfix in graph_kernel/splitter.py, the kernel includes InplaceAssign should be a composite node. 4. added two tool functions GetAllInputDeviceTypes and GetAllOutputDeviceTypes into AnfAlgo. 5. do not fuse a single Assign in pass BasicOpsFusion.	5 years ago
Gaoxiong	32e19e83da	update graph kernel split model for Ascend	5 years ago
dayschan	7599686a72	GraphKernel supports multi-output kernels	5 years ago
dayschan	37a48f6aac	GraphKernel supports GPU 1. Update akg submodule 2. Refactor akg_kernel_build, akg_ascend_kernel_build, akg_gpu_kernel_build 3. Add akg_kernel_json_decoder to support converting kernel_json to AnfNode. 4. Add GraphKernel Cost Model. (mindspore/_extends/graph_kernel) 5. Add some GraphKernel passes to GpuSession, move these passes to backend/optimizer/graph_kernel. 6. Add global id for ir files. 7. Fix bug in ConstInputToAttr.	5 years ago